Vol. 133:287-297, 1996 MARINE ECOLOGY PROGRESS SERIES

Mar Ecol Prog Ser Published March 28

Effects of high-molecular-weight dissolved organic matter on nitrogen dynamics in the

Mississippi River plume

Wayne S. Gardner1·*, Ronald Benner2, Rainer M. W. Amon2 , James B. Cotner, Jr3,

Joann F. Cavaletto1, Jeffrey R. Johnson4

1NOAA Great Lakes Environmental Research Laboratory, 2205 Commonwealth Blvd, Ann Arbor, Michigan 48105, USA 2Marine Science Institute, University of Texas at Austin, Port Aransas, Texas 78373, USA

3 Department of Wildlife and Fisheries Sciences, Texas A&M University, College Station, Texas 77843, USA 4 Cooperative Institute for Limnology and Ecosystem Research, University of Michigan, Ann Arbor, Michigan 48105, USA

ABSTRACT: The dynamics of N and its interactions with labile dissolved organic C (DOC), bacteria, and phytoplankton were studied to determine potential effects of dissolved organic matter (DOM) and light on N dynamics in surface waters of the Mississippi River (USA) plume in the Gulf of Mexico. Bac­terial uptake of added labeled N compounds e5NH4 +or 15N-labeled dissolved free amino acids, DFAA) was stimulated more by high-molecular-weight (HMW, > 1 kDa) DOM than by low-molecular-weight (LMW, < 1 kDa) DOM. An index that inversely indicated the presence of labile DOC was defined as the fraction of assimilated Amino acid- 15~ that was Recovered as 15N-.8,mmonium (ANRA), following the additions of high-levels (4 llM) of 15N-DFAA. ANRA ratios were high in the absence of other available carbon sources because heterotrophic bacteria were forced to use the added amino acids as a carbon source for respiration rather than as a nutrient source for biomass formation. In dynamic light/dark experiments, conducted with in situ populations of organisms, uptake rates of added 15NH4 +were sig­nificantly enhanced both by the presence of light and by the addition of HMW DOM. Uptake rates of added 15N-labeled DFAA were increased by the addition of HMW DOM but not by light. ANRA ratios were consistently lower in the presence of added HMW DOM than in controls. Added HMW DOM thus appeared to stimulate the incorporation of assimilated DFAA into bacterial biomass. Bacterial growth rates were relatively high in both light and dark botUes with DFAA additions and in light bottles with HMW DOM plus NH/ additions, but they remained comparatively low in dark bottles with added NH4 • . These results are consistent with the idea that bacterial N dynamics in these euphotic waters may be tightly coupled to photosynthetic activities over short time scales.

KEY WORDS: Dissolved organic carbon · Nitrogen cycling · Ammonium · Bacteria · Phytoplankton · Amino acids

INTRODUCTION

A large fraction of organic material from primary production in aquatic ecosystems is thought to cycle through dissolved organic matter (DOM) and subse­quently through heterotrophic bacteria (Scavia & Laird 1987, Chin-Leo & Benner 1992, Chr6st & Rai 1993), but biogeochemical mechanisms responsible for these interactions in surface waters are not yet well defined

'E-mail: gardner@glerl.noaa.gov

© Inter-Research 1996 Resale of full article not permitted

(Munster & Chr6st 1990). The bulk of the DOMin the ocean is considered to be quite resistant to bacterial breakdown (Menzel 1974 and references therein) but a significant fraction of the DOM in surface waters can be asssimilated or respired by bacteria within days (Kirchman et al. 1991). Bacterial incubations with isolated DOM that had been fractionated into high­molecular-weight (HMW, >1 kDa) and low-molecular­weight (LMW, < 1 kDa) components indicated that HMW DOM supported much more bacterial growth and respiration than LMW DOM (Amon & Benner

288 Mar Ecol Prog Ser 133: 287-297, 1996

1994). These results seem to argue against the para­digm Lhal bacterial growth in natural waters is primar­ily supported by LMW DOM, but they could be biased by the fact that fluxes of LMW DOM in the water are not necessarily reflected by their concentrations in the water at the time of DOM isolation. If labile LMW DOM compounds are removed from the water as rapidly as they are produced, concentrations could be low in isolated fractions even though their fluxes are quantitively important to bacterial/microbial foodweb dynamics under in situ conditions.

Chemical analysis of HMW DOM isolated from ocean water indicates that it has a high concentration of carbohydrates and C:N ratios of about 15 (Benner et al. 1992). Similarly, HMW dissolved organic carbon (DOC) from Mississippi River (USA) plume surface waters have C:N ratios ranging from 14 to 20 as com­pared to ratios of 19 to 26 in the river (R. Benner unpubl. data). These ratios are much higher than ratios for bacterial biomass, suggesting that the isolated labile HMW DOM may be an important carbon source for bacteria (Sakugawa & Handa 1985, Pakulski & Ben­ner 1994) but less important as a N source. Bacteria that grow on this material therefore must obtain most of their N from inorganic sources or from LMW organic N compounds that are not retained by Lhe ultrafiltra­tion process (Amon & Benner 1994).

Recent studies with added 15N have demonstrated that under natural light, significant dissolved organic nitrogen (DON) is released from phytoplankton (Bronk & Ghbert 1993). This recently released DON may be assimilated or metabolized by bacteria (Keil & Kirch­man 1991, 1993, Simon & Rosenstock 1992) or possibly by phytoplankton (Palenik & Morel 1990a, b) andre­incorporated into the food web via the microbial loop (Bronk et al. 1994). These apparent differences in the dynamics of DOC and DON may be explained by a partial chemical uncoupling of DON and DOC as microbial substrates (Kirchman et al. 1991). For exam­ple, rapid cycling of labile, photosyn-thetically produced LMW DOM, e.g.

these incubations by the absence of light and by the removal of organisms larger than bacteria. In the sec­ond 'natural-biota' experiments, we conducted light and dark 15N isotope addition experiments on unfil­tered surface waters in the presence and absence of added HMW DOM that had been isolated from sea­water at the site. These experiments included natural organisms and allowed photosynthesis and associated DOM production/microbial interactions to continue in the lighted bottles.

Comparison of results from these 2 types of experi­ment provides insights about the dynamic role of HMW DOM and light-driven photosynthetic processes in the short-term turnover of inorganic and organic N compounds. In this paper, we specifically consider the following questions for surface waters in the Missis­sippi River plume: (1) is HMW DOM a source or sink for dissolved inorganic N compounds? (2) Do bacteria use different organic substrates for growth (biomass formation) and energy (metabolism) in surface waters? (3) Are bacterial production and heterotrophic N cycling rates directly enhanced by photosynthetic pro­duction of available DOC or DON?

METHODS

Study sites and DOC analysis. Experiments were conducted on board the RV 'Longhorn' in July 1993 in the northern Gulf of Mexico in the vicinity of the Mis­sissippi River plume. Samples were collected at 2 sites (Table 1) of intermediate salinity where surface-water primary production (Lohrenz et al. 1990), bacterio­plankton production (Chin-Leo & Benner 1992), and nutrient cycling rates (Cotner & Gardner 1993) are high relative to corresponding rates in surrounding waters. DOC was measured in treatment-bottle waters with a Shimadzu TOC 5000 analyzer (Benner & Strom 1993).

DFAA, may be an important compo­nent of DON turnover (Fuhrman 1990, Kirchman et al. 1990).

Table 1. Sample site locations and surface-water characteristics for the fraction­ated-DOM-bacterial (FDOMB) and natural-biota experiments. Note, the experi­mental precision of the DOC measurements is less than 2% coefficient of variation

To gain further insights about N transformations mediated by primary producers and heterotrophic bacteria, we conducted 2 types of experiment: in the first 'fractionated-DOM-bacter­ial' experiments, we examined micro­bial-nitrogen interactions with HMW and LMW DOM, respectively, that had been isolated from the water at a dis­crete time point. Autotrophic DOM production was prevented during

Sampling date

Sampling coordinates

Temperature (0 C}

Salinity (psu)

DOC cone. ()lg-atom C 1-1)

FDOMB experiment

July 14, 1993

29° 10.3'N, 91°29.8'W

30

8 LMWDOM: 175 HMWDOM:269

Natural-biota experiment

July 23, 1993

28°49.5'N, 92°05.84'W

30

24 No added HMW DOM: 246

With added HMW DOM: 525

Gardner et al. : Effects of organic matter on nitrogen dynamics 289

Nitrogen transformation experiments. In the 'frac­tionated-DOM-bacterial' experiment, HMW (> 1 kDa) DOM was physically separated from LMW (<1 kDa) DOM, diluted back to natural concentrations in artifi­cial seawater, and incubated in the dark as previously described (Amon & Benner 1994). Seawater was fil­tered through a 0.1 pm pore-size hollow-fiber filter to remove particles, and the DOM was partitioned into HMW and LMW fractions by tangential-flow ultra­filtration (Benner 1991). A 2.5 1 concentrate of HMW DOM was then diluted to natural concentrations with ca 15 1 of artificial seawater. LMW DOM remained in the original seawater after ultrafiltration. Each fraction was inoculated with a natural bacterial assemblage (<0.6 pm) that had been isolated from the initial water sample. The DOC concentrations in incubations with HMW and LMW DOM were 269 and 175 pM C 1-1

,

respectively (Table 1). Isotope dilution and enrichment experiments, with

15NH4 +or 15N-labeled amino acid additions to the frac ­tionated-DOM treatments, were conducted in the dark at in situ temperatures. After natural bacteria had been added to both treatments (Amon & Benner 1994), sub­samples (70 ml) from each were placed in 75 ml tissue culture bottles and treated with 4 pM levels of either 15NH/ or 15N -labeled DFAA (MSD Isotopes MN-2625 Algal Amino Acid Mixture). Each isotope-addition experimental treatment was run in triplicate bottles for experimental replication. Ammonium and DFAA con­centrations and ammonium isotope ratios W5NH/]: (total NH4 +])were monitored a few minutes after addi­tion of the labeled compounds and at 2 or 3 additional intervals over 20 to 24 h.

'Natural-biota' isotope dilution and enrichment experiments were conducted similarly in a shipboard incubator both under natural light and in the dark with natural seawater (controls) and with natural seawater treated with concentrated HMW DOM isolated from the same sites (Table 1). An assumption for these experiments is that NH4 + can be assimilated by either phytoplankton or bacteria, whereas DFAA are more likely to be assimilated by bacteria (Wheeler & Kirch­man 1986). Relative bacterial growth rates were mea­sured after incubations of ca 24 h to provide insights about the interactive effects of added HMW DOM, N (in the form of NH4 + or DFAA), and light on bacterial growth rates.

To measure dissolved N~ + and DFAA concentra­tions and (15NH4 +]:(total NH4 +] ratios, 10 ml of water were sampled from each treatment and passed through a 0.2 pm pore-size nylon filter (25 mm diame­ter). The first 3 rnl was used to rinse the filter and the next 7 rnl was collected in a clean 8 ml vial (Wheaton # 224884) . Ammonium and DFAA concentrations were measured on-board ship (Gardner & St. John 1991).

and the remaining filtrate was frozen for later isotope ratio analysis by high performance liquid chromatog­raphy (HPLC; Gardner et al. 1991, 1993). Community ammonium regeneration and potential uptake rates in the bottles were calculated from changes in dissolved ammonium concentrations and isotope ratios over time using the Blackburn-Caperon model (Blackburn 1979, Caperon et al. 1979).

Calculation of the percentage of assimilated Amino acid-15N Recovered as 15N-Ammonium (ANRA) in DFAA addition experiments. Isotope enrichment experiments with relatively high-level additions of 15N-labeled DFAA can provide useful information on the 'maximum uptake velocity' Wmaxl of DFAA by the bacterial community (Wright & Hobbie 1965) and can also provide a qualitative indication of the relative availability of other naturally occurring labile DOC substrates to the bacterial community (see discussion below). These experiments differ from typical radioac­tive tracer studies of amino acid dynamics in that they are used as an indicator of the presence or supply rates of labile organic carbon rather than as a tool to trace the in situ dynamics of natural amino acids in the water. The use of 'saturating' rather than 'tracer' con­centrations of DFAA minimizes isotope dilution of the added DFAA with natural DFAA and allows estimation of Vmax for DFAA in the water. In addition, the fate of 15N rather than of 14C or 3H from the added labeled amino acids is followed. The concentrations of amino acids (4 11M) that were added are high relative to nat­ural concentrations of amino acids, but are reasonable, relative to potential turnover rates, for these summer experiments in Mississippi River plume surface waters (Cotner & Gardner 1993).

The ANRA ratio provides an index that can be assumed to be inversely related to the amounts or sup­ply rates of non-amino labile organic compunds that are available to bacteria if effects of bacterial grazers and uptake of ammonium by phytoplankton are accounted for. If the added amino acids (C:N ratio = 4.30) become the predominant forms of carbon that are available to bacteria, a high percentage of the amino acid carbon will be metabolized, to meet respiration/ deamination energy requirements, and a correspond­ingly high percentage of the 15N added as amino acids will in turn be mineralized to ammonium (e.g. Zehr et al. 1985, Goldman et al. 1987). For example, in the Mis­sissippi River plume region, about 90% of the 'assimi­lated' 15N that had been added as labeled amino acids was recovered as 15NH4 + in bottle experiments with dark subsurface waters as compared to about 48% in surface waters under natural light (Gardner et al. 1993). Similarly, organic substrate-depleted Lake Michigan bacteria regenerated an amount of ammo­nium approximately equivalent to the amount of added

290 Mar Ecol Prog Ser 133: 287-297, 1996

DFAA consumed (Gardner et al. 1987, 1989). Con­versely, if other sources of labile carbon are available to be metabolized, more of the amino-N can be incor­porated into biomass (see 'Results').

ANRA, expressed as %, was calculated as follows:

where (NH4 +It is the concentration of NH4 + at the sam­pling point where DFAA had reached background (or near-background) levels, R1 is the isotope ratio of (

15NH4 +]:[total NH4 +] at the same sampling point that [NH4 +]1 was determined, [NH4 +],is the concentration of NH4 + at the initial sampling point, Ri = isotope ratio of [15NH/):(total Nil/) at the initial sampling point, (AAJ1

is the measured DFAA concentration at the initial sam­pling point, and (AA]r is the measured amino acid con­centration at the sampling point where DFAA concen­trations had reached near-background levels. In using this equation, we assume that changes in DFAA levels are caused by uptake of the added 15N-labeled amino acids. If initial concentrations of DFAA are higher than 4 J..lM plus the background, i.e. if measurable natural DFAA are present in the water in addition to the back­ground fluorescent response typically observed for non-labile compunds (Gardner & St. John 1991, Keil & Kirchman 1993), the decreases can exceed 4 pM due to the presence of non-labeled labile amino acids. When measurable unlabeled DFAA were present in any of the treatments, we assumed that (AA); - [AAJ1 = 4 pM for the calculation, since the change in 15N-labeled amino acids cannot exceed the amount that was ini­tially added. The background response was equivalent to that from DFAA levels of 0.4 and 0.8 pM, respec­tively, in waters used for the fractionated-DOM-bacte­rial and natural-biota experiments. Note, ANRA ratios should be considered as a qualitative rather than an absolute indicator of substrate availability because their values can also be affected to some extent by other factors such as isotope dilution of the DFAA pool, mineralization of assimilated amino acids by bacterial grazers, ard/or uptake of 15NH4 + by phytoplankton in experiments with natural assemblages of organisms. To minimize the effects of these factors, we calculated ANRA at the first point where DFAA were reduced to near-backuround levels and conducted some experi­ments in the dark as well as under natural light condi­tions.

Bacterial abundances, growth rates, and amino acid turnover rates. Bacterial abundances were estimated with acridine orange staining and epifluorescence microscopy of duplicate samples (Hobbie et al. 1977). Relative bacterial production rates were estimated from rates of incorporation of 3H-thymidine into DNA (Fuhrman & Azam 1982). Samples were incubated in duplicate \vi.th 20 nM (final concentration) of [methyl-

3H]thyrnidine (TdR) (Amersharn). Nucleic acid and protein fractions were separated with a trichloroacetic acid extraction procedure (Chin-Leo & Benner 1992). Killed blanks were used to correct for abiotic adsorp­tion to filters. In killed controls, formalin was added to a final concentration of 4%, 15 to 20 min prior to the addition of radioisotope.

Amino acid turnover rates were estimated with a mixture of 3H-labeled DFAA from algal protein hydrolysate (Amersham). Tracer amotmts (less than 1 nM final concentration) were added to duplicate samples and incubated for 10.0 min at ambient tem­perature. Incubations were terminated by filtration onto a 0.2pm pore-size filter (Millipore GSWP). Dupli­cate filter and filtrate samples were collected and frozen prior to further analyses. Filters were dissolved with 1 rnl ethylacetate before radioassay. Turnover rates were corrected for respired amino acids by allow­ing a 1 rnl sample to equilibrate with 2 rnl of distilled water through the gaseous phase by a modification of the procedure of Ashcroft et al. (1972). The filtrate of a labeled sample was preserved (2% formalin final con­centration) and placed in a 1.5 ml open centrifuge tube inside of a closed 20 ml scintillation vial containing distilled water. The respired tritium eH-H20 ) was allowed to equilibrate at 25°C for 1 mo. Samples from the inner and outer vials were analyzed in a liquid scintillation counter. All dpm (disintegrations per minute) in the large vial were assumed to be from 3H-H20. The fraction of amino acids respired was cal­culated as (dpm0 w)/(dpmp + dpmF), where dpm0 w. dpmp, and dpmF are dpm in distilled water, on the filter and in the filtrate, respectively. All samples were cor­rected for background levels of 3H-H20 by analysis of killed controls. Background levels were less than 5% of the total dpm. Incorporation was calculated as dpmp/(dpmp + dpmp), and efficiency was estimated as Incorporation/(Incorporation +Respiration).

RESULTS

Fractionated-DOM-bacterial experiment

Except for an increase at the second sampling point, the concentrations of added NH4 +remained relatively constant in the seawater containing LMW DOM in the 15NH4 +treatments (Fig. 1). However, in the water con­taining HMW DOM, NH4 +was removed to near-back­ground levels during the 20 h incubation. This result is consistent with the hypothesis that HMW DOM is an 'available' organic substrate with a high C:N ratio that causes microbes to assimilate NH4 +. In contrast, LMW DOM must have either been unavailable or had a rela­tively low C:N ratio so that the bacteria did not require

Gardner et al.: Effects of organic matter on nitrogen dynamics 291

6 HMW +15NH/ LMW +15NH/

1.0

~ 0.8 .2: 4 !I! c 0.6 0 ·~ + 0.4 '<I' c 2 I (]) z (.) 0.2 c Iii 0 0 0 0.0 0 '0 HMW +15N·AA LMW +15N-AA

t:.. '() <( 6 1.0 + 0 '<I'

I c 0.8 z .E 4 1.{)

<( 0.6--'0 .Q c 0.4 Cii C1l

2 a: + '<I'

I 0.2 z

0.0 5 10 15 20 0 5 10 15 20 25

Time (hours)

Fig. 1. Time-course results for ammonium (e ) and amino acid (0) concentrations and 1 5NH4~ isotope ratios (• ) for the frac­tionated-DOM-bacterial experiment with HMW and LMW DOM treatment waters treated with either 15NH4 • or 15 N-

labeled amino acids and incubated for 20 h

additional N. Removal patterns for 15N-labeled DFAA were similar in both treatments. After an initial lag, probably due to relatively low bacterial numbers, removal rates were high; the added DFAA (4 pM) were completely removed within about 20 h (Fig. 1). The ANRA ratios were higher in the LMW DOM treatment (40%) than in the HMW DOM treatment (22%) where NH4 + concentrations did not appreciably increase (Table 2).

For the LMW DOM treatment, NH4 + regeneration rates decreased from about 60 nM h -I in the first 2 in­vervals to approximately 0 in the third interval (Fig. 2). This trend may have reflected the depletion of labile organic nitrogen. Ammonium regeneration rates for the HMW DOM treatment varied among replicate measurements and did not show significant changes

Table 2. Summary of ANRA values ± SE (n in parentheses) expressed as a percentage for the different amino acid addi­tion treatments in the the fractionated-DOM-bacterial (FDOMB) and natural-biota experiments. See text for ANRA

calculation procedure

Dark Light

HMW LMW

FDOMB 22 ± 3 (3) 40 ± 2 (3)

HMW Control HMW Control Natural-biota 36 ± 2 (2) 65 ± 2 (3} 27 ± 3 (3) 61 ± 4 (2)

'.c 012

:2 0. 10 ..5 .2! m

0.08

a: 0.06 c: 0 0.04 ~ Q) c: 0.02 Q) 0> 0.00 Q)

a: ·0.02

0.4

'.c 0.3 :2 ..5 0.2 Q)

.>< 0.1 m a. :::J 0.0 • ~ ·0.1 c Q)

0 ·0.2 Cl..

·0.3 0 4 8 12 16

Time (hours)

Fig. 2. Mean potential uptake a nd regeneration rates for the ammonium-addition experiments during incubations for HMW DOM (D) and LMW DOM (e) treatments calculated from data presented in Fig. 1. Rates were calculated from changes in ammonium concentrations and isotope ratios

using the model of Blackburn (1979)

between initial and final time points. Calculated up­take rates for NH/ ranged from <0 during the first in­terval for both treatments, when bacterial abundances were relatively low, to about 0.2 pM h- 1 for the second interval and to 0.3 pM h- 1 for the HMW DOM al the third interval (Fig. 2). Uptake rates decreased after 6 h for the LMW DOM treatment but continued to increase during the second interval in the HMW DOM treat­ment, reflecting changes in bacterial nutritional status for the 2 types of treatment (Amon & Benne.r 1994).

Relative bacterial growth rates and amino acid incor­poration and respiration rates, determined with tracer levels of 3H-labeled DFAA (Cotner & Gardner 1993}, were measured after incubating the fractionated mate­rial for 1 d. Bacterial abundances and thymidine incor­poration rates were both substantially higher in the HMW DOM treatments than in the LMW DOM treat­ments (Fig. 3a) . Amino acid incorporation and respira­tion rates, and amino acid incorporation efficiencies, were all higher in the HMW DOM treatments than in the LMW DOM treatments (Fig. 3b}.

Natural-biota experiment

Ammonium was removed fastest from the lighted bottles with added HMW DOM and slowest from con­trol water in the dark (Fig. 4}. More NH4 +was removed

292 Mar Ecol Prog Ser 133: 287-297, 1996

~

'..c 25 ~ B c 20 .Q ~ 15 0 e-0

10 0 ..!: CD c 5 '0 .E >- 0 ..c f-

0.7

0.6

'..c 0.5

c 0

0.4

t5 ~

0.3 u_

0.2

0.1

0.0

A

LMW

B

Lcl LMW

Am1noac•d Respiration Incorporation

HMW

25 E .!:!2

20 Qi 0

<D 0 ,...

15 CD 0 c

10 ., '0 c ::::)

5 .0 <(

:-ffi 0 (i;

0 ., m

Fig. 3. (A) Bacterial abundances and thymidine incorporation rates and (B) amino acid respiration and incorporation rates and calculated incorporation efficiencies (= Incorporation/ (Incorporation+ Respiration)) for 3H-labeled amino acids afte r treatment waters from the fractionated-DOM-bacterial expe r-

iment were incubated for about 24 h in the dark

from solution in bottles containing HMW DOM and in lighted bottles than in control seawater treatments. The same trends were seen in Isotope ratios indicating isotope dilution of 15 NTT/ with 14NH/ in all of the treatments

Ammonium regeneration and potential uptake rates were calculated over the first 10 h because changes over the first 2.5 h interval were not sufficient to pro­duce an adequate signal-to-noise ratio to allow recog­nition of significant trends. A 2-way ANOVA, testing the effects of light (p = 0.13) and HMW DOM addi­tions (p = 0.22) over the 10 h interval, did not show significant differences in NH4 + regeneration rates. The HMW dark treatment showed lower regeneration

DARK NATURAL LIGHT 8 1.0

6 0.8

4 ...... ..____ ----. 0.6

0.4

~ 2 0.2

~ 0 0.0 c +HMW+NH

4•

.r 0 0.8

'<t :.= 6 I ~ 0.6 z c 4 I§ Q) 0.4 (.) 0 c 2 +HMW +NH/ 0.2 t:. 0 (.) {)

0.0 .r "0 0 "<<' '(3 CONT.+AA I

<( 6 0.8 z

0 "' 0.6 ,...

c '-'

E 4 .Q <( 0.4 1U "0 2 0.2 a: c <U 0 0.0

+ "<<'

I 0.8 z 6 0.6

4 0.4 2 ·-0.2

0 0.0 0 5 10 15 20 0 5 10 15 20 25

Time (hours)

Fig. 4. Time-course results for ammonium (e ) and amino acid (O) concentrations and 1 'NH4 • isotope ratios (• J for the nat­ural-biota experiment conducted in the dark and under nat­ural light and in the presence and absence of added HMW DOM. Treatment waters were fortified with either 15Nll4 • or

15N-labeled amino acids and incubated for 23 h

rates than the light treatments but SEs for the other treatments overlapped (Table 3). In contrast, uptake rates during the same interval showed significant dif­ferences for both light (p = 0.0013) and HMW DOM additions (p = 0.019) with progressive increases from the dark control to the light HMW addition treatment (Table 3). The presence of natural light had a slightly greater effect than the addition of HMW DOM but results from the 2 treatments were generally additive,

i.e. interactions between the 2 treatments

Table 3. Ammonium regeneration and uptake rates (pM Nl V h 1) ± SE in natural-biota experiments for Control and HMW treatment bottles incu­bated in the dark and under natural light conditions. Note, these data were calculated from the ammonium-addition experiments presented in Fig. 4

were not significant (p = 0.59). Interest­ingly, in the dark control treatments, uptake and regeneration rates were not significantly different from each other,

RegeneraLon rate Uptake rate

Dark

Control HMW

0.08 :t 0 .04 0.05 ± 0.002 0.09 :t 0.04 0.18 ± 0.01

Light

Control HMW

0.13 ± 0.02 0.09 ± 0.03 0.23 ± 0.02 0.29 ± 0.02

but in all other treatments potential uptake rates were significantly higher than the regeneration rates (Table 3) .

Removal rates for added DFAA, which reflected the Vmax of the bacterial commu­nity for DFAA, were high, particularly in

Gardner et al.: Effects of organic matter on nitrogen dynamics 293

the HMW DOM treatments where DFAA were com­pletely removed in about 10 h (Fig. 4). By comparison, more than 20 h was needed for complete removal of the added DFAA in the controls. The fact that HMW DOM addition immediately enhanced the DFAA removal rates (i.e. during the first interval) suggests that the native bacterial populations were already adapted to using DFAA at relatively high rates in the presence of natural HMW DOM. The DFAA removal patterns were similar in light and dark bottles for both control and HMW DOM treatments.

Isotope ratios of NH4 + generally increased in pro­portion to decreases in 15N -labeled DFAA concentra­tions. However, NH4 + accumulation from DFAA degradation was minimal in the treatments with added HMW DOM. More NH4 + accumulated in the dark than in the light and in the controls than in the waters treated with HMW DOM (Fig. 4) . The ANRA ratios were about 36 and 26% in the dark and light HMW DOM treatments as compared to 64 and 61% in the corresponding control treatments (Table 2). In the treatments with added HMW DOM, the isotope ratios increased proportionately with the decreases in DFAA concentrations, but then decreased slightly after the DFAA were depleted to background levels (Fig. 4). This decrease can be attributed to isotope dilution of the 15NH4 +that was in solution after degra­dation of the 15N-labeled DFAA. Final NH4 + concen­trations were similar in comparable treatments whether NH/ or DFAA was the initial source of avail­able N (Fig. 4).

Effects of light and HMW DOM additions on relative bacterial growth rates

Relative bacterial growth rates were measured in all of the above treatments that had been enriched with 15NH/ or 15N-labeled DFAA and incubated for about 24 h under light (natural) or dark conditions (Table 4). The highest growth rates were observed in bottles incubated in the light with added HMW DOM and DFAA. Similar high growth rates were observed in lighted bottles containing HMW DOM

DISCUSSION

Is HMW DOM a source or sink for dissolved inorganic nitrogen compounds?

In agreement with expectations from the chemical analyses showing a high C:N ratio and carbohydrate content for HMW DOM (Benner et al. 1992, Pakulski & Benner 1994), our data suggest that HMW DOM is a source of C for heterotrophic bacteria that in turn form a biological sink for NH4 +. Bacterial abundances and growth rates increased much more in the presence of isolated HMW DOM than in the presence of natural levels of LMW DOM (Amon & Benner 1994). In all of our experiments with HMW DOM, NH4 + concentra­tions decreased more in the presence of added HM\1\T DOM than they did in its absence. The contention that HMW DOM is an available C source for bacteria is supported by the fact that it caused increased uptake of both forms of N in the natural-biota experiments but did not enhance NH4 + regeneration rates.

The presence of HMW DOM in experiments with DFAA additions greatly enhanced the Vmax of the bac­terial community for DFAA uptake and consistently caused ANRA ratios to be lower than was observed in comparable control bottles without HMW DOM addi­tions. In the fractionated-DOM-bacterial experiment, removal patterns for 15N -labeled DFAA were similar in both treatments but ANRA ratios were higher with LMW DOM (40%) than with HMW DOM (22%). The HMW DOM apparently enabled more of the DFAA-N to be incorporated into biomass than did the LMW DOM. Thymidine incorporation rates and DFAA incor­poration and respiration rates were all higher ir1 the HM\1\T DOM treatments than in the LMW DOM treat­ment. The incorporation efficiency (ratio of DFAA incorporation rate: DFAA incorporation rate+ respira­tion rate) was also slightly higher for the HMW DOM than for the isolated LMW DOM. These results are consistent with the idea that HMW DOM is a bioavail­able substrate with a relatively high C:N ratio (Amon & Benner 1994) but did not provide evidence for high reactivity of isolated LMW DOM.

and added NH4 +. Much lower rates were observed in the dark bottles with NH4 +

additions (with or without added HMW DOM) and in the light treatment without HMW DOM. In contrast, relatively high

Table 4. Relative bacterial growth rates (pM TdR h-1) ± SE in control and HMW natural-biota experiments incubated in lhe dark and under natural light conditions with 4 1-1M additions of 15NH4 + or 15N-labeled amino acids.

growth rates, comparable to results for light bottles, were observed in dark bot-tles with added DFAA. Incubations in the light with added NH4 + and HMW DOM produced approximately the same result as adding the DFAA directly.

See Fig. 4 legend for information on sample treatments

+ ISNH/ + 15N-labeled amino acids

Dark Control HMW

33 ± 8 106 ± 10

26 ± 9 125 ± 14

Light

Control HMW

47 ± 12 92 ± 9

131 ± 32 156 ± 2

294 Mar Ecol Prog Ser 133: 287-297, 1996

In our natural-biota experiment, ANRA in the dark was higher in the control treatments (64%) than in the ones treated wi lh HMW DOM (36%). Both of th ese percentages were higher than the corresponding con­versions observed in the fractionated-DOM-bacterial experiment. This difference could result from sam­pling site differences or more likely from the fact that remineralizing rnicrograzers were present in the natural-biota experiment but were removed before incubations in the fractionated-DOM-bacterial experi­ment. Likewise, in the light ANRA was higher in the control (61 %) than in lhe HMW DOM treatment (26%). The observation that ANRA was consistently higher in control treatments than in the HMW DOM ones suggests that HMW DOM enhanced incorpora­tion of amino acid nitrogen into bacterial biomass (vs respiratory loss of the organic nitrogen as NH4 +). The similarity of ANRA values in comparable light and dark treatments (Table 2) indicates that a relatively small portion of the mineralized ammonium was taken up by phytoplankton before ANRA measure­ments were made.

Do bacteria use different organic substrates for growth and energy in surface waters?

The experiments described here in conjunction with the corresponding data of Amon & Benner (1994) sug­gest that the dynamics of DOC and DON are partially uncoupled (Kirchman et al. 1991). The HMW DOM that can be isolated by ultrafiltration consists largely of carbohydrates and has a relatively high C:N ratio (Ben­ner et al. 1992). This fresh material appears to be an important source of C for bacteria but does nol appear to provide sufficient N for bacterial growth. Thus, N must come from inorganic forms or from other sources such as DFAA or peptides (Coffin 1989, Keil & Kirch­man 1991, Rosenstock & Simon 1993). Of course, the dynamics of DOC and DON cannot be completely uncoupled if labile LMW DON compounds such as DFAA or peplides are quantitatively important metabolites because these compounds contain both organic C and N.

Studies examining the fate of 15NH4 + and 15N03 in isotope addition experiments suggest that a substantial fraction of the 15N incorporated into organisms under natural light is released as 15N-DON with relatively short turnover times (Bronk & Glibert 1993, Bronk el al. 1994). This DON release can result as a by-product of feeding by zooplankton or protozoans (Nagata & Kirchman 1991, Bronk & Glibert 1993) that tend to release HMW DON, or from passive release of LMW DON by autotrophs (I lellebust 1974, Bronk & Glibert 1993). Our results in combination with previous analy-

sis of HMW DOM (e.g. Benner et al. 1992) suggest that any labile DON that is produced must occur mainly as short-lived LMW DOM. If substantial DON is released as proteins that are stable for more than a few hours, one would expect IIMW DOM isolated by ultrafiltra­tion to have a lower C:N ratio than has been observed (Benner et al. 1992). Likewise, HMW DOM would tend to be a source rather than a sink for nitrogen in dark bottle incubations (Amon & Benner 1994).

Are bacterial production rates and heterotrophic N cycling rates directly enhanced by photosynthetic

production of available DOC or DON?

Bacterial numbers and growth rates (Amon & Benner 1994) and NH4 + uptake rates were directly enhanced by the presence of fresh HMW DOM. The rapid removal of DFAA in both the HMW and LMW treat­ments in the fractionated-DOM-bacterial experiment suggests that the Vmax for DFAA uptake was high, as would be expected at the experimental temperature of 30°C. As mentioned above, only about 22% of the assimilated 15N from the DFAA was recovered as 15NH4 + in the HMW DOM treatment as compared to 40% for the LMW DOM treatment. The presence of HMW DOMas a C source apparently allowed the bac­teria to use the DFAA for biomass incorporation more efficiently than when added DFAA alone were the main source of carbon.

The fractionated-DOM-bacterial experiment did not show greatly enhanced microbiological activity or N cycling rates in the presence of isolated LMW DOM. This observation would seem to argue against the par­adigm that microbial processes are driven by LMW organic compounds. However, another feasible expla­nation for this observation is that bacterial activity decreases when bacterial processes are experimen­tally uncoupled from autotrophic processes that serve as short-term sources of labile organic materials to the bacteria. Recent studies have shown a close relation­ship between bacterial production and photosynthetic processes (Cole et al. 1982, 1988, Coffin et al. 1994). If turnover of the labile organic compounds is rapid, and if bacterial uptake rates are as rapid as production rates of the labile compounds, steady-state concentra­tions of some materials (e.g. DFAA) remain very low even though the actual fluxes of the materials are high. Our natural-biota experiments allowed comparison of N cycling rates during limes when active autotrophic production of bacterial substrates should have occurred (light experiments with natural biota) to those when new production of labile organic substrates should have been low (fraclionated-DOM-bacterlal and dark natural-biota experiments).

Gardner et al.: Effects of organic matter on nitrogen dynamics 295

Our ANOVA results indicate that the presence of both light and added HMW DOM increased NH4 +

uptake and isotope dilution rates. Light had a slightly greater effect on NH4 + uptake rates than added HMW DOM but the effects were comparable and non-inter­active. These results suggest that both bacteria and phytoplankton were actively assimilating NH4 + in the light with added HMW DOM.

The minimal differences between light and dark DFAA removal patterns that were observed both in the presence and absence of added HMW DOM indi­cate that heterotrophic bacteria rather than phyto­plankton were primarily responsible for DFAA uptake. The high removal rates (i.e. Vmaxl for the added DFAA, particularly in the presence of added HMW DOM, suggest that the bacteria may have been accustomed to turning over relatively large quantities of DFAA under natural conditions even though DFAA concentrations are normally low. The HMW DOM apparently increased the efficiency of DFAA uptake by heterotrophic bacteria as DFAA uptake rates approximately doubled when isolated HMW DOM was added. This increase in DFAA uptake rates was already observed during the first 4 h incubation inter­val, an indication that the bacteria were likely already adapted to using this combination of substrates (i.e. DFAA with HMW DOM). The ANRA values were consistently lower in HMW DOM treatments than in controls, suggesting that the HMW DOM stimulated the incorporation of amino acid-N into bacterial bio­mass. A conceptual model that would explain these observations is that DFAA, produced by autotrophic processes, are rapidly used by heterotrophic bacteria mainly for biomass formation, particularly when labile HMW DOM with a high C:N ratio is available as a source of energy for respiration.

Examination of bacterial growth rates in the differ­ent 15N-addition treatments provides revealing insight about the effects of DFAA and autotrophic production processes on bacterial production in these waters. Addition of DFAA stimulated growth rates in all treat­ments as may be expected if DFAA are a preferred source of N by bacteria (e.g. Kirchman et al. 1989, Keil & Kirchman 1991). The addition of HMW DOM appeared to stimulate bacterial growth rates in the presence of added DFAA in the light and dark and in the presence of NH4 + in the light but not in the dark. Lowest bacterial growth rates were observed for NH4 +

additions in the dark and only slightly higher growth rates were observed for NH4 + additions in the light treatment without added HMW DOM. We unfortu­nately did not measure bacterial production rates in comparable bottles without N additions and therefore could not determine the extent that NH4 + addition stimulated bacterial production rates over those with-

out nitrogen additions. However, our results show that NH4 + additions to HMW DOM in the light can clearly stimulate bacterial production more than the same additions in the dark.

The increased bacterial production rates in the lighted HMW DOM treatment suggest that autotrophic processes interacted with HMW DOM to enhance bac­terial growth rates. Possibly, there was autotrophic conversion of the 15NH4 +to readily available organic N compounds such as DFAA in the presence of light and a source of labile carbon. Alternatively, the phyto­plankton may have provided the bacteria with other 'growth factors' that allowed them to efficiently assim­ilate NH /in the presence of HMW DOM. These data are interesting relative to diel differences in 'bacterial growth capacity' observed for bacteria in microcosms (Zweifel et al. 1993). The growth capacity, defined as the growth yield of bacteria in filtered water relative to natural bacterial abundances in the same water, tended to be lower at night than during the day (Zweifel et al. 1993). Also, in nutrient-enriched meso­cosms, bacterial production, 0 2 consumption, and growth efficiencies were highest during daylight or early evening when substrate availability was highest (Coffin et al. 1994). In Chesapeake Bay, amino acid uptake rates were also higher in daylight than at night (Glibert et al. 1991).

The above experiments suggest that nutrient cycling processes in Mississippi River plume surface waters are closely linked to light-driven organic mat­ter production or conversion by the photosynthetic process. The ANRA index appears to effectively indi­cate the relative degree that labile DOC was available to support bacterial respiration in the different treat­ments. HMW DOM appears to be a more important source of C and energy for bacteria than LMW DOM. In contrast, LMW DOM may be an important N source for bacteria, as suggested by high v;nax values for DFAA and increased incorporation rates of high levels of DFAA into biomass in the presence of HMW DOM. Our data agree with the hypothesis that HMW DOC compounds provide a carbon source to 'fuel' bacterial growth but that rapidly-recycling LMW organic N compounds (e.g. DFAA), or possibly some other growth factors plus NH4 +, are needed for bacterial biomass formation in surface waters of the Mississippi River plume.

Acknowledgements. This research was sponsored by the NOAA Coastal Ocean Program through lhe Nutrient Enhanced Coastal Ocean Program (NECOP). We thank the crew of the RV 'Longhorn' for ship support, Lynn Herche for statistical analyses, and Harvey Bootsma, Peter Landrum, and David Kirchman for constructive comments on the manu­script. This paper is GLERL Contribution No. 937.

296 Mar Ecol Prog Ser 133: 287-297, 1996

LITERATURE CITED

Amon RMW, Benner R (1994) Major role for high-molecular­weight dissolved organic matter in the marine carbon cycle. Nature 369:549-552

Ashcroft SJH, Weerasnghe LCC, Bassett JM, Randle PJ (1972) The pentose cycle and insulin release in mouse pancreatic islets. Biochem J 126:525-532

Benner R (1991) UltrafiltratiOn for the concentration of bacte­ria, viru~es, and dtssolved organic matter. In: Hard DC, Spencer DW (eds) Marine particles: analysis and charac­terization. Geophysical Monograph 63, American Geo­physical Union, Washington, DC

Benner R, Pakulski JD, McCarthy M, Hedges Jl, llatcher PG (1992) Bulk chemical characteristics of dissolved organic matter in the ocean. Science 255:1561-1564

Benner R. Strom M (1993) A critical evaluation of the analyti­cal blank associated with DOC measurements by high temperature catalytical oxidation. Mar Chern 41:1 3

Blackburn HT (1979) Method for measuring rates of NH4 •

turnover in anoxic marine sediments, using a 1sN-NH4 •

dilution technique. Appl Environ Microbiol37:760-765 Bronk DA, Glibert PM (1993) Contrasting patterns of dis­

solved organic nitrogen release by two size fractions of estuarine plankton during a period of rapid NH4 • con­sumption and N02 production. Mar Ecol Prog Ser 96: 291-299

Bronk DA. Glibert PM, Ward BB (1994) Nitrogen uptake, dis­solved organic nitrogen release, and new production. Sci­ence 265:1843-1846

Caperon J, Schell, D, Hirota J, Laws E (1979) Ammonium excretion rates in Kaneohe Bay, Hawaii, measured by a 15N isotope dilution technique. Mar Biol54:33-40

Chin-Leo G, Benner R (1992) Enhanced bacterioplankton pro­duction at intermediate salinities in the Mississippi River plume. 1\lar Ecol Prog Ser 87:87-103

Chr6st RJ, ~ai H (1993) Ectoenzyme activity and bacterial secondary production in nutrient impoverished and nutn­ent enriched freshwater mesocosrns. Microb Ecol 25: 131-150

Coffin RB (1989) Bacterial uptake of dissolved free and com­bined amino acids in estuarine waters. Limnol Oceanogr 34:531-542

Coffin RB, Connolly JP, llarris PS (1994) Availability of dis­solved organic carbon to bacterioplankton examined by oxygen utilization. Mar Ecol Prog Ser 101:9-22

Cole JJ, Findlay S, Pace ML (1988) Bacterial production in fresh and saltwater ecosystems: a cross-system overview. Mar Eco. Prog Ser 43:1 10

Cole JJ, Likens GE, Strayer DL (1982) Photosynthetically pro­duced dissolved organic carbon: an important carbon source for planktonic bacteria. Limnol Oceanogr 27: 1080-1090

Cotner JB, Gardner WS (1993) Heterotrophic bacterial medi­ation of ammonium and dissolved free amino acid fluxes in the Mississippi River plume. Mar Ecol Prog Ser 93: 75-87

Fuhrman JA (1990) Dissolved free amino acid cycling in an estuarine outflow plume. Mar Ecol Prog Ser 66:197-203

Fuhrman JA, Azam F (1982) Thymidine incorporation as a measure of heterotrophic bacterioplankton production m marine surface waters: evaluation and field results. Mar Bioi 66:109-120

Gardner WS, Chandler JF. Laud GA, Carrick HJ (1987) Sources and fate of dissolved free amino acids in epi­limnetic Lake Michigan water. Limnol Oceanogr 32: 1353-1362

Gardner WS, Cotner JB Jr, Herche LR (1993) Chromato­graphic measurement of nitrogen mineralization rates in marine coastal waters with 15N. Mar Ecol Prog Ser 93: 65-73

Gardner WS, Chandler JF, Laird GA (1989) Organic mineral­ization and substrate limitation of bacteria in Lake Michi­gan. Limnol Oceanogr 34:478-485

Gardner WS, Herche LR, St. John PA, Seitzinger SP (1991) High performance liquid chromatographic determination of 15NH4 : (

14NH4 + "NH4 Jton ratios in seawater for isotope dilution expenments. Analyt Chern 63:1838-1843

Gardner WS, St. John PA (1991) High-performance liquid chromatographic method to measure ammonium ion and primary amines in seawater. Analyt Chern 63:537-540

Glibert PM, Garside C, Fuhrman JA, Roman MR (1991) Time­dependent coupling of inorganic and organic nitrogen uptake and regeneration in the plume of the Chesapeake Bay estuary and its regulation by large heterotrophs. Lim­no! Oceanogr 36:895-909

Goldman JC, Caron DA, Dennett MR (1987) Regulation of growth efficiency and ammonium regeneration in bacteria by substrate C:N ratio. Limnol Oceanogr 32:1239 1252

Hellebust JA (1974) Extracellular products. In: Stewart WOP (ed) Algal physiology and biochemistry. Univ of California Press, Berkeley, p 838-863

Hobbie JE, Daley RJ, JasperS (1977) Use of Nuclepore filters for counting bacteria by fluorescence microscopy. Appl Environ Microbiol33:1225-1228

Keil RG, Kirchman DL (1991) Contribution of dissolved free amino acids and ammonium to the nitrogen requirements of heterotrophic bacterioplankton. Mar Ecol Prog Ser 73: l-10

Keil RG, Kirchman DL (1993) Dissolved combined amino acids: chemical form and utilization by marine bacteria. Limnol Oceanogr 38:1256- 1270

Kirchman DL, Keil RG, Wheeler PA (1989) The effect of amino acids on ammonium utilization and regeneration by het­erotrophic bacteria in the subarctic Pacific. Deep Sea Res 36:1763-1776

Kirchman DL, Keil RG, Wheeler PA (1990) Carbon limitation of ammonium utilization and regeneration by hetero­trophic bacteria in the subarctic Pacific. Lim no! Oceanogr 35:1258-1266

Kirchman DL, Suzuki Y, Garside C, Ducklow HW (1991) lligh turnover rates of dissolved organic carbon during a spring phytoplankton bloom. Nature 352:612-614

Lohrenz SE, Dagg ML, Whitledge TE (1990) Enhanced pri­mary production at the plume/oceanic interface of the Mississippi River. Cont Shelf Res 10:639-664

Menzel DW (1974) Primary productivity, dissolved and partic­ulate organic matter, and the sites of oxidation of organic matter. In: Goldberg ED (ed) The sea, Vol5, Marine chem­istry. Wiley, New York, p 659- 678

Munster U, Chr6st RJ (1.990) Origin, composition and micro­bial utilization of dissolved organic matter. In: Overbeck J, Chr6st RJ (eds) Aquatic microbial ecology. Biochemical and molecular approaches. Springer, New York, p 8-46

Nagata T, Kirchman DL (1991) Release of dissolved free and combined amino acids by bacterivorous marine flagel­lates. Limnol Oceanogr 36:433-443

Pakulski JD, Benner R (1994) Abundance and distribution of carbohydrates in the ocean. Limnol Oceanogr 39:930 -940

Palenik B, Morel FMM (1990a) Amino acid utilization by marine phytoplankton: a novel mechanism. Limnol Oceanogr 35:260- 269

Palenik B, Morel FMM (1990b) Comparison of cell-surface L-amino acid oxidases from several marine phytoplank­ton. Mar Ecol Prog Ser 59:195-201

Gardner eta!.: Effects of organic matter on nitrogen dynamics 297

Rosenstock B, Simon M (1993) Use of dissolved combined and free amino acids by planktonic bacteria in Lake Con­stance. Limnol Oceanogr 38:1521-1531

Sakugawa H, Handa N (1985) Chemical studies on dissolved carbohydrates in water samples collected from the North Pacific and Bering Sea. Oceano! Acta 8:185- 196

Scavia D, Laird GA (1987) Bacterioplankton in Lake Michi­gan: dynamics, controls, and significance to carbon flux. Lirnnol Oceanogr 32:1017-1033

Simon M, Rosenstock B (1992) Carbon and nitrogen sources of planktonic bacteria in Lake Constance studied by the composition and isotope dilution of intracellular amino

This article was presented by B. & E. Sherr (Senior Editorial Advisors), Corvallis, Oregon, USA

acids. Limnol Oceanogr 37:1496-1511 Wheeler PA, Kirchman DL (1986} Utilization of inorganic and

organic nitrogen by bacteria in marine systems. Limnol Oceanogr 31:998-1009

Wright RT, Hobbie JE (1965) The uptake of organic solutes in lake water. Linmol Oceanogr 10:22-28

Zehr JP, Axler RP, Goldman CR (1985) Heterotrophic mineral­ization of amino acid nitrogen in a subalpine Castle Lake, California. Mar Chern 16:343-350

Zweifel UL, Norrman B, Hagstrom A (1993) Consumption of dissolved organic carbon by marine bacteria and demand for inorganic nutrients. Mar Ecol Prog Ser 101:23- 32

Manuscript first received: June 12, 1995 Revised version accepted: October 4, 1995